Using a BMS as a remote relay/contactor

[convme]

Hi,

I've bought a BMS without low temperatur cut off and found a way to control this through the use of external controllers.

My problem is, I would need an expensive normally closed (nc) relay to get it all work. This would cut off the current under certain circumstances. I don't want to go more into details, because this is just the backend story and should not be part of the discussion here.

The idea is to use a smaller and cheaper NC-relay to control the BMS which would then act like a big relay.

I've realized, the B- to P- is normally open when the BMS is not powered. Giving power to it this contact closes. That's exactly what I need.
Now I could attach a NC-relay between the thin B1- line which would cut off the BMS power and force the main-contact to open, and the other way round.

I've heard, this might be dangerous for the BMS, and I'd like to know if someone already made experiences here.



There are some multichannel-relays out there, maybe it would be less dangerous to cut off and reconnect all the 5 lines at the same time, or just the B*+ lines? (https://www.aliexpress.com/item/32649659086.html?spm=a2g0s.8937460.0.0.6f4b2e0esufgKd)

 

[JeepHammer]

So what you are trying to do is shut down the BMS instead of using a high amp contactor to shut the entire battery off from charging/load?

 

[FilterGuy]

Hi,

I've bought a BMS without low temperatur cut off and found a way to control this through the use of external controllers.

My problem is, I would need an expensive normally closed (nc) relay to get it all work. This would cut off the current under certain circumstances. I don't want to go more into details, because this is just the backend story and should not be part of the discussion here.

The idea is to use a smaller and cheaper NC-relay to control the BMS which would then act like a big relay.

I've realized, the B- to P- is normally open when the BMS is not powered. Giving power to it this contact closes. That's exactly what I need.
Now I could attach a NC-relay between the thin B1- line which would cut off the BMS power and force the main-contact to open, and the other way round.

I've heard, this might be dangerous for the BMS, and I'd like to know if someone already made experiences here.

View attachment 6298

There are some multichannel-relays out there, maybe it would be less dangerous to cut off and reconnect all the 5 lines at the same time, or just the B*+ lines? (https://www.aliexpress.com/item/32649659086.html?spm=a2g0s.8937460.0.0.6f4b2e0esufgKd)

I do like the 'out of the box' thinking. Sometimes that leads to great ideas.

Like you, I have heard that cutting the small B- while other things are connected can damage the BMS. I do not have personal experience either way with that. Furthermore, this is going to be dependent on how the BMS is designed. (One BMS may be fine with the disconnect and another may blow up.)

In another thread (that I have not been able to find) a poster said he tried disconnecting the small B- on his BMS and it did not turn off the BMS. However, that is also going to be BMS specific.

Just thinking about what must be going on inside the BMS, it *may* be more likely that the most positive lead would turn off the BMS. Here is my thinking:
The BMS circuitry must have a Power + and Power - to opperate, If you cut either the power + or Power - to the circuitry, it should shut off. Furthermore it would be unusual for the design to use anything but the most positive voltage for it's power line. Since there is only one connection to B+, it is not a big leap to assume the BMS get's it's power from B+. Meanwhile, the BMS has two connections to B- so it is less clear where the Power- is connected.

However, even if it does not damage the BMS and does turn off the BMS, be careful about adding resistance to the lead. There is a widely held belief that it is important to keep the leads from a BMS all the same length in order to keep the resistance the same for all of the leads. I have yet to hear a clear explanation for why this is critical (and how critical it is), but I have no way to prove it wrong either.

 

[JeepHammer]

However, even if it does not damage the BMS and does turn off the BMS, be careful about adding resistance to the lead. There is a widely held belief that it is important to keep the leads from a BMS all the same length in order to keep the resistance the same for all of the leads. I have yet to hear a clear explanation for why this is critical (and how critical it is), but I have no way to prove it wrong either.

I'm not sure I believe that...
Looks like common wire to me, and several BMS boards without leads say to use common 22ga (or whatever) wire,

Others have excessively long leads and tell you to cut to length for your Series connections.
Not a one I've looked at endlessly for the last 5-6 weeks says a thing about using any particular material for series connections (copper, aluminum, nickel alloy, etc) and about all show the leads connecting to the connector bar and not the battery terminals.

I'd probably use something like communication wire, high conductivity and low resistance/impedance wire if I had to add wire, but I haven't had to do it... Yet...
So it's speculation through observation on my part having not had to spend hours on customer service phone lines or weeks waiting on emails to get returned with answers to questions.
(We all know how that goes...)

What is for sure, the BMS won't allow charge/discharge when the leads disconnect.
YouTube videos show 4 or 5 people, Including Will Prowse, finding disconnected leads from BMS and the battery is effectively dead.

That's one thing...
Connecting or disconnecting UNDER LOAD might very well be something else entirely.
I can see a big load, or charge waiting when BMS connects/disconnects being a big issue,
But since I haven't done it I just don't know what will actually happen...

Maybe get one of those $20 BMS units and try it is the only way I know...

 

[FilterGuy]

My problem is, I would need an expensive normally closed (nc) relay to get it all work. This would cut off the current under certain circumstances.

I know you are hesitant about Aliexpress, but...https://www.aliexpress.com/item/33001613800.html

 

[FilterGuy]

I'm not sure I believe that...
Looks like common wire to me, and several BMS boards without leads say to use common 22ga (or whatever) wire,

Others have excessively long leads and tell you to cut to length for your Series connections.
Not a one I've looked at endlessly for the last 5-6 weeks says a thing about using any particular material for series connections (copper, aluminum, nickel alloy, etc) and about all show the leads connecting to the connector bar and not the battery terminals.

I'd probably use something like communication wire, high conductivity and low resistance/impedance wire if I had to add wire, but I haven't had to do it... Yet...
So it's speculation through observation on my part having not had to spend hours on customer service phone lines or weeks waiting on emails to get returned with answers to questions.
(We all know how that goes...)

I hear ya..... Like I said, it seems to be one of those 'facts' that everyone treats as gospel but no one can back up. In a different thread I have tried modeling the circuit and I could not find a defensible argument that the length makes any significant difference, but that only means I failed to prove a problem. It does not mean a problem does not exist. I would love someone to explain it with something more than 'you must'. However, I figured I would bring it up, along with doubts, so the OP could decide for himself (Herself?).

 

[JeepHammer]

I just ordered that contactor! Thanks for the link.
Since I'm using used cells, I *Shouldn't* (not can't) parallel cells, that means 4 times the cell strings and 4 times the disconnects.
200A on each cell string, 400A to disconnect the bank entirely, I *Should* never draw more than about 150A at 48V, but you just never know...

 

[JeepHammer]

I hear ya..... Like I said, it seems to be one of those 'facts' that everyone treats as gospel but no one can back up. In a different thread I have tried modeling the circuit and I could not find a defensible argument that the length makes any significant difference, but that only means I failed to prove a problem. It does not mean a problem does not exist. I would love someone to explain it with something more than 'you must'. However, I figured I would bring it up, along with doubts, so the OP could decide for himself (Herself?).

Not enough of that...
It's 'Word From GOD' or nothing most times.
It's not as clear cut as someone trying to run 100A through a 16Ga wire!
(Smoke & panic follow shortly! )

I'm not sure too many people have a multimeter with enough places behind the decimal point to check the resistance in 8" of common communications wire...

I'm tempted to order one of those $20-$30 BMS's and see if this works myself, it would save big, heavy duty contactor costs, kicker relays to cycle those contactors, and 'Creative' wiring with all those connections/moving parts that can go south...
Particularly if the BMS is programmable for high/low voltage and charge/discharge rate protections.

 

[convme]

I know you are hesitant about Aliexpress, but...https://www.aliexpress.com/item/33001613800.html

Yeah, I though of this one, too... or at least a similar, but somehow I was confused about the technical specs telling me, the auxiliary contact could only be at 24VDC or 48VDC.

I also got into trouble with the limitations of the battery-case dimensions. Now even the BMS needs a breakthrough in one of the outer walls to make it all fit inside. Which will result in having the heatsink outside. The good point here is, no more unwanted heat inside the case during the summer and everyone can see what BMS is installed. The bad point, I'm not sure, if there still is enough place inside the case for a big relay. So, here's an other reason why to try out the BMS thing as a relay/contactor.

 

[convme]

Connecting or disconnecting UNDER LOAD might very well be something else entirely.
I can see a big load, or charge waiting when BMS connects/disconnects being a big issue,
But since I haven't done it I just don't know what will actually happen...

Maybe get one of those $20 BMS units and try it is the only way I know...

That's exactly the reason why I haven't tried it out yet preferring to ask here first.

My BMS was a little bit expensive to do these kind of experiments. Maybe if one could approach starting with lightweight loads and rise by round, than maybe I could risk placing the bet on the table

I'm tempted to order one of those $20-$30 BMS's and see if this works myself, it would save big, heavy duty contactor costs, kicker relays to cycle those contactors, and 'Creative' wiring with all those connections/moving parts that can go south...
Particularly if the BMS is programmable for high/low voltage and charge/discharge rate protections.

Cool thingI'd appreciate this for sure, although we'd have to wait a while until it arrives.

 

[JeepHammer]

Yeah, I though of this one, too... or at least a similar, but somehow I was confused about the technical specs telling me, the auxiliary contact could only be at 24VDC or 48VDC.

I also got into trouble with the limitations of the battery-case dimensions. Now even the BMS needs a breakthrough in one of the outer walls to make it all fit inside. Which will result in having the heatsink outside. The good point here is, no more unwanted heat inside the case during the summer and everyone can see what BMS is installed. The bad point, I'm not sure, if there still is enough place inside the case for a big relay. So, here's an other reason why to try out the BMS thing as a relay/contactor.

Either you are the mechanical/electrical DIY type or not... Its something that either clicks in your head or it doesn't.
*IF* it does, then you have an accumulation of tools & parts/'Junk' and if you are lucky, you have a workbench.

What I did was parts carts under my workbench, battery strings on carts, Anderson connectors to take the strings in and out of 'Bank'.
The workbench protects cells in bank, and I get a workbench!
Pegboard & shelves go up around, or even over charge controllers, inverters, etc on the wall behind the workbench.
Pegboard still works well on hinges so you can get to that stuff.

I'd rather be wasting time 'Tinkering' at my workbench than wasting time in front of the 'Boob Tube' loosing IQ points... I don't care what celebri-tard is doing or wearing today...


I hang both manual disconnects (Anderson Connectors),
And automatic disconnectors (contactors) on battery strings In Bank, and a contactor on the 'Main Lines'/Buss from Battery Bank to inverter.
The only thing that has direct access to the mains/buss is the emergency system, fire detection/supression & system shutdowns.

I'll sacrifice battery power to save as much of the system as possible, and since it's only activated when there is a BIG emergency (like a fire) it's Normally Closed (NC), takes power to open the safety relays/contactors.
Just a few Watts a month for sensors until it's tripped in an emergency, but when there is an extra emergency who cares about battery power when the entire system, and maybe the house is going up?
I sure don't!

A little bit of trivia, 48V contactor coils are usually activated at 36 to 60V or more.
If the coil gets warm with prolonged use, hang a resistor in the line to the contactor coil.
I *Expect* to pull 150A at the absolute maximum, 48V nominal battery system,
Everything I'm using is rated for 200A minimum and 60V minimum.

The main lines/busses are 400A contactor, 400A cables, 400A terminals, etc.
Even though they *Should* never see more than 150-160A.

Everything, and I mean EVERYTHING, gets 'Last Line' fuses.
The 150A battery strings get cables for 250A cables & 200A fuses.
I can over-run 'Nominal' by a full 1/3 before the fuse gives up, 2/3 before the cables/terminals heat up.
Keep in mind a 250A cable doesn't instantly turn into a ball of flame at 251A, it just starts to heat a little, it would take 450-500A to make it heat enough to catch the insulation on fire or melt the insulation...
I have a LOT of safety built in, and this is from experience, I have removed my eyebrows before, and I'm not eager to repeat that stupidity.
I also sleep where my system is, so safety is a big deal when you aren't even awake...

The other thing I do is 'Noid' lights, 5¢ or 10¢ worth of resistor and LEDs.
Green LEDs when things are powered and working the way they should be, Red when something fails.
For 10¢ I don't have to spend half a day collecting tools and troubleshooting before I get an idea of where the fault/fail is.
YEARS of building machines with remote sensors taught me that, some as long as a building so finding one bad contact/sensor/switch can take days...
Just look for the 'Red Light' and you have the problem.

While everyone wants to rush to solid state...
Electromagnetic relays *May* fail more often (on safety systems they only activate when someone is bad wrong, so almost no use/fails) but the fail is obvious and easy to find.
Not many transistors or MOSFETs come with 'Noid' lights, it you want them you have to add them anyway...
A Noid light is the ONLY way to see if a MOSFET is working or not, while you can watch, or feel the vibration of a mechanical relay.

I won't use a proximity sensor without a Noid light, since you can't see a proximity sensor work, no sense in getting one without a light...

It's up to everyone to do what they think is best, but for me green lights mean 'Working', Red lights mean 'Failure'.
For the cost of a tiny resistor and LED, it's just too stupid simple for me NOT to do.

 

[convme]

Hmm, I don't know if i can follow this all, but by now everything is still in 3D development, so there's nothing going on making something at the workbench yet. I'm also still missing some parts that are either not bought yet or in shipment. Many dimensions I got are just taken from the technical specs of the products.

This is a capture of the current development state (without descriptions of the single parts):


As you can see, there is not much space left inside, I could add anything more. And I'm still missing all the cables here. The outer dimensions of the case are very limited, and after the last revision I made there's not even 1mm left to grow anymore. That's the reason, why the big red block in the front has to break through the wall. It was intended originally to become installed vertically in the front left corner, but it did not match from the height because of the contacts I didn't model and who simply need more space than expected. The good thing is, now I have more space in the front to probably install a relay and even the space to install a second BMS in parallel below the first one (All the other selected components are prepared to handle far more than 200A so I could easily double the current). But by now, 200A has to be enough.

I'm still not sure if I even want to have a relay but if so, I would prefer a SSR although I'm not sure yet if it would waste less space than the classical relays when it comes to the cooling that has to be installed, too.

 

[JeepHammer]

Hmm, I don't know if i can follow this all, but by now everything is still in 3D development, so there's nothing going on making something at the workbench yet. I'm also still missing some parts that are either not bought yet or in shipment. Many dimensions I got are just taken from the technical specs of the products.

This is a capture of the current development state (without descriptions of the single parts):
View attachment 6642

As you can see, there is not much space left inside, I could add anything more. And I'm still missing all the cables here. The outer dimensions of the case are very limited, and after the last revision I made there's not even 1mm left to grow anymore. That's the reason, why the big red block in the front has to break through the wall. It was intended originally to become installed vertically in the front left corner, but it did not match from the height because of the contacts I didn't model and who simply need more space than expected. The good thing is, now I have more space in the front to probably install a relay and even the space to install a second BMS in parallel below the first one (All the other selected components are prepared to handle far more than 200A so I could easily double the current). But by now, 200A has to be enough.

I'm still not sure if I even want to have a relay but if so, I would prefer a SSR although I'm not sure yet if it would waste less space than the classical relays when it comes to the cooling that has to be installed, too.

Is your cooling required because of charge/discharge rates or something else?
Since I'm temperature controlled (earth sheltered) I've considered a water bath for cooling.
I used a water bath way back when I first started with flooded lead acid and the container was above ground, the sun got it so hot it melted plastic stuff inside.

I always see people pack the batteries as close together as they can, but when mine were at the mercy of weather, I spaced them apart in summer, close them up and insulated them in winter.
Now I'm using aluminum dividers in the cases as heat sinks, and with reasonable charge/discharge rates I'm not seeing a lot of temp related problems.

Right now, under .2 C charge/discharge rate, I'm more interested in longevity that big power numbers.

 

[convme]

Is your cooling required because of charge/discharge rates or something else?
Since I'm temperature controlled (earth sheltered) I've considered a water bath for cooling.
I used a water bath way back when I first started with flooded lead acid and the container was above ground, the sun got it so hot it melted plastic stuff inside.

There are not temp problems yet, and there also is no need for a cooling.

When it comes to the BMS, the Daly BMS has two heatsinks installed by default, one from each side. One of them now is outside the battery-box. But I did it because of space-problems, not because of cooling problems. It's just a nice additional feature to have the heat emitted outside the battery-box, at least partially.

When it comes to the SSR, I just read about big SSR with heavy loads need some further heatsinks to become cooled properly.

My battery cells are very close, that's right, but these are aluminium-cells and I have a spacer between every one although it's very slim and you can't see it in the capture.

I used a water bath way back when I first started with flooded lead acid and the container was above ground, the sun got it so hot it melted plastic stuff inside.

?

EDIT: When the right time comes, I'll post all the details about the project and its development inside the "show and tell" area.